Sensible Digital Transformation in Manufacturing with Dave Griffith at the Manufacturing Hub Podcast (Season 2, Episode 8)

• Why Dave is so passionate about manufacturing.
• Why is manufacturing the “wild west,” according to Dave Griffith?
• The different factors driving digitization in manufacturing.
• What is “sensible digital transformation”? And why is it so important to be sensible?
• Are people unrealistic about the effort involved in digital transformation?
• How simulation and digital twins can drive digital transformation.
• How machine learning and AI can help with sensible digital transformation.

Dave Griffith:

I think that the biggest change that we’ve seen over the course of the last few years, we see a lot of people going and retiring, and I think when we talk about that, we talk a lot about the workforce, but what I’ve seen a lot of people is ownership and leadership of these companies, right? Lots of the ownership-leadership of these organisations are getting to the point of, “Hey, I have been unwilling to invest my time and money in automation or people or any of these other things that we’re talking about. I think it’s time to retire.” So I see lots of the next generation of people coming up, and the next generation of people are absolutely more than willing to go and invest in the technology.

Chad Ghalamzan:

Welcome to the Engineering Innovation podcast, powered by Simcenter. This podcast is a show by engineers, for engineers, in which we explore how engineering simulation and tests are contributing to a healthier, happier, and more sustainable future. Our guests bring in-depth knowledge of their specialty, and we have engaging conversations about the past, present, and future of engineering.

My name is Chad Ghalamzan. My guest today is Dave Griffith, co-host of the Manufacturing Hub podcast, which aims to inform and inspire leaders in manufacturing automation and related fields. He’s here to share some of his experience and knowledge with us today. Hello, Dave. Welcome to the show. Why don’t you start by telling us a little bit more about you?

Dave Griffith:

Absolutely. Chad, thank you so much for having me here. As I like to say, my name is Dave Griffith. I have been in manufacturing automation for the last 15-plus years. Throughout the course of that, I’ve had the opportunity to do a bunch of different things. My technical background is in aerospace, and so I’ve got to work within aerospace. I spent some time working for large OEMs that build the large entry-style machines, that drill and and rivet aeroplane fuselages together.

And just a bit of a teaser: I can’t tell you how much I wish I had simulation technology back when I was doing that. The last eight years or so of my career have been fairly split. I spent a chunk of that running a systems integration company. We were focused mostly on the MES, Manufacturing Execution Systems, level. So we did a lot of work with data, a little work with simulation.

And then, the last four years, in addition to being the co-host of Manufacturing Hub, I have run my own company called Kaplan Solutions. We focus on the execution problems. We focus on going and taking data if it exists, and if it doesn’t exist, understanding what the core problems are that exist at end-user facilities, finding ways to go fix the problems, such as downtime and quality and throughput and on-time delivery.

Chad Ghalamzan:

You mentioned that you wished you had engineering simulation, so let’s maybe start with, what did you have then? What was your initial work experience like then?

Dave Griffith:

Absolutely. I’ve spent a lot of time in my career working for small companies, and anyone who has spent time working with small companies, integrators, groups like that. I call it a combination of a blessing and a curse. I was very blessed to be able to do any and everything, but it was also a curse of it’s like, “Hey, Dave, you’re in your early twenties. Do you want to do this?”

And one of those asks for me that I said yes to, and I’m very happy I did at that point, was, “Hey, Dave, we’ve got to go build a facility that is in theory greenfield, and we’re going to go build the last couple of pieces of the fuselage. We’re going to go build the riders.” The potential end user said, “Hey, we won this bid. Go build us a facility.” It was amazing. And I spent the next two months of my life in a borrowed conference room with basically everything that you could think of that you need to go put aircraft together.

So I worked for a company that the core of what they did is they build the large gantry-style machines that drill and rivet the aeroplane fuselages together. So that is a couple hundred feet long, give or take, and we’re fifty, maybe a hundred, feet wide, depending upon what is going on. We’ve got lots of, at that point in time, KUKA robotic arms. Of course, the controllers were not tight enough for aircraft tolerances.

So the mad geniuses at this organisation were literally ripping controllers off of these robots, putting brand-new, custom-made controllers on there, so we could go get the tolerances correctly. We had almost football field-sized cells that we had. And then it’s like, so we have this, but where are we putting the raw materials? And so you go figure out vertical warehousing. And 12, 15 years ago, vertical warehousing wasn’t as easy to go Google online to go figure out what that looks like, but we’ve got vertical warehousing.

And then, beyond vertical warehousing, it’s like, okay, we’ve got to go move these things from cell to cell. So it’s like we’ve got to go figure out the UGVs, AGVs in order to go figure out how are we going to go move these pieces? There’s going to be some amounts of manual rework. At the end point of that, it’s going to go somewhere. We’re going to QC test it, and then we’ve got to go wrap all of this up because, of course, it couldn’t be simple.

So we’re basically going to put these in sea containers, we’re going to put them on a ship, and we’re going to go ship them somewhere else. And so we had all of these pieces literally printed out on pieces of paper, taped up against the wall. And then, the thing that I remember maybe most fondly is the entire conference room table was just full of engineering graph paper.

It was cut out in theory to in scale of what we wanted this half-a-million-square-foot facility to look like. And then we took and we cut engineering graph paper or we shrunk down the sizes of these UGVs and the fuselages and everything like that. And we would physically go take our fingers and go run them around throughout the facility to see, “Hey, what is going to hit? And if it is going to hit, what do we have to do? Do we need to give ourselves more room?”

And Chad, every time I see what we can do with simulation now, I see lots of simulation in automotive, I see tonnes of simulation opportunities in other places. It’s like, man, if I had to go design another aircraft facility, or really any facility, you know who’s not going to borrow a 25-foot conference room table and go do it all manually by hand? Dave. Dave is going to go find someone much smarter than I am to go simulate this, and I don’t know, save 30 or 50 or 500 hours of my life.

And then, on the back end, one of the things that I love simulation for, and we’ll get more to this later, is the fact that seeing is believing. So some of us are very blessed of going and being able to understand how everything is going to move. I like to, for non-manufacturing folks, compare it to those house renovation shows. Some people can walk into a house and only see what is there.

Some people can walk into the house and, in their brain, they can go take out a wall, they can redo the kitchen, they can redo the bathroom, they go mock it up and go help other people see and understand. For me, simulation is the, “we’re going to go mock it up and see and help people understand what is going on.” So if I were to go do something like this again, not only do I get the raw engineering benefit of, “Hey, we are going and showing you that what we’ve designed is one, well thought through, and two, we’re not going to have issues running into everything.”

But I’m able to go into a conference room, into a boardroom and go sell the project. And these are very expensive projects, right? Tens or hundreds of millions of dollars of projects. We had a bunch of mad scientists PowerPoint at that point who’s like, “This is how we’re going to raise and lower everything. This is how we’re going to move it.” And it was all tiny blocks across 500 to 700 PowerPoint slides of “we’re going to go ahead and do this.”

And as I look at simulation, it’s, “I’m going to go press play, and we’re going to go watch everything move around in an, I don’t know, a three-minute video, a five-minute video.” And seeing is believing. Most of us are very visual people. Most of us don’t want to read 500 pages of documentation when a 90-second video is going to go solve the problem. So for me, simulation solves those problems.

We all, as engineers, all us folks in manufacturing know rework is one of the dirtiest words we can possibly say, right? Because we all want to pretend that we’re all good a hundred percent of the time, when we all know we’re not really good a hundred percent of the time. So, we realised that we had missed this rework station. And so, there were a lot of long nights and weekends, ordered-in Chinese food of, “Hey, we’ve got to go rip all of this apart, and we’ve got to go find space for 10 rework stations throughout the process of this.”

And again, it’s one of those that doing it by hand is more valuable than not doing it by hand. But if we have to go through the process of choosing, do I do it all on paper by hand or can I go simulate this? Again, simulation is from my side as I like to tell people on manufacturing up, I would not do it again by hand.

It was a good experience. It’s like calculus or all of those other things. It’s in theory good to know how to do it by hand, but man, I’m really happy I can go have a computer programme or calculator any of these things that can go solve some of these problems for me.

Chad Ghalamzan:

What makes you so passionate about manufacturing?

Dave Griffith:

That is a loaded question, Chad, and I appreciate that. I guess at the core, I have looked at a number of different things throughout the course of my career. First and foremost, manufacturing continues to pull me back in. I went to go look at going into aerospace and going to do that. And the first great opportunity that I had was the company that I just mentioned. It’s aerospace. I can use all of my technical knowledge of what this is supposed to look like, but we’re building aeroplanes and as soon as you get into it, and for me as soon as I got into it at a medium and a high level, I like to define manufacturing as the Wild West. And it’s one of my Americanist.

Chad Ghalamzan:

Why the Wild West? Is it because it’s unregulated, unlawful? What makes it so much the Wild West?

Dave Griffith:

It is all of those things, Chad, because in the outside, where we are today, it is 2023. We all have these super computers that are the smartphones in our pockets. Half of us are wearing smart watches. It feels like the cars today, that we buy today basically go drive themselves. And we walk in to a manufacturing facility, half of the things are from 1954.

I had a client, great family, but he liked to joke the first time I met him in twice a month for the next four years, he would joke, “Hey, it might be 2020 or 2023 outside, but as soon as you walk through those gates, it’s 1954.” It is true. But in that instance, all the guys were in suits and ties in the office, everyone working on the floor was all in their matching uniforms. And beyond that, so much of the technology is 1950s, 1960s, 1980s.

Chad Ghalamzan:

Maybe you could explain for people because I wasn’t around in 1954. So when you say it was like 1954, just maybe give us a quick idea. What did it look like?

Dave Griffith:

Back then we had lots of relays, we had lots of timers, a tonne of things. The vast majority of them were all manual processes. And when we talk about smart factories, when we talk about industry 4.0, digital transformation, and I certainly talk about all of those things. We talk as to what they could be in the future. But as a practitioner, I go walk out on the floor, and we are struggling with basic automation.

We’re struggling with automation, we’re struggling with sensors, we’re struggling with basic data collection on so many different groups and so many different places. And so, almost every place you go to, they are running into very similar processes, but we’ve got the opportunity to make generationals with so many of these different facilities.

You go walk in, you see lots of PLC5s. I was talking to a group last week that had PLC2s, which I have never seen running, but apparently, this beer facility is running a PLC2, and it’s running the entirety of the beer facility. And so, I call it the Wild West because there are just so many opportunities and every place you go, there are so many opportunities. We kinda publicise the groups that do a really good job, and everyone thinks, “Hey, we might be so far behind because all of these places are doing so well.” In reality, there are a couple of places that are doing really well, but the vast majority of the groups still have 1970 to 1990s technology that they are just absolutely running on, and they continue to run on it because it works.

Chad Ghalamzan:

Do you think it’s been cost? Do you think it’s been the difficulty to switch over? Why have they stayed frozen?

Dave Griffith:

That’s a good question. In my opinion, and I guess I spent a number of years trying to go sell and convince groups of upgrades. I am 95% convinced that they run it because it still works. And it is difficult to go look at spending $50,000 or even $10,000 on an upgrade, plus taking something down for potentially weeks at a time, because it still works.

And at the core of manufacturing, like the business case of most of these manufacturers, is we have to make cases or we have to have pallets or we have to make product, and we are judged upon the product that we put out the door. And that is why so many of us run to failure. We run to failure because that’s the way that we’ve always done it. And forward-thinking is great, but we spend so much of our time fighting fires in the now at this moment, it’s really difficult to consider what are we going to do six months from now? What are we going to do two years from now?

I have talked to so many engineering and other managers that are like, “This solution that we’re talking about, will it run for three years?” And we’re like, “Absolutely. It’ll run for three years. It’ll run for a lot longer than three years.” And they look at me, Chad, and they’re like, “I don’t care. I’m retired in three years. After three years, it’s someone else’s problem.” So we very much intrinsically have the, if it’s my problem, I’m going to focus on it. If it’s not my problem, I’m just going to punch it down the road and let someone else figure it out.

Chad Ghalamzan:

So the vision is maybe not there in terms of what the potential could be as long as it’s running today, that seems to be satisfactory to a lot of facilities. Besides the lack of innovation that’s been brought into these facilities, what else do you see as a place for them to continue to grow? Or what else about it makes you so passionate?

Dave Griffith:

A couple of things. Technology on the operational technology side, it could be newer hardware, it could be newer software, it could be data analytics. It could be all of these things that we seemingly take for granted, again in our everyday life, but it’s also workforce and workforce enablement. And I talked a bit about execution and one of the passions, one of the things that I’ve been working on is execution.

For me, I find that it’s amazing that groups continue to run into similar processes. Almost every manufacturer that I talk to nearly across the world, has the same problems in 2023. We’ve got the problem of, it is difficult to hire workers. And then if I can hire workers, it’s really difficult to train and retain those workers, but they continue to do the exact same thing that they’ve done forever.

We go hire someone, maybe there’s an onboarding process, and then basically, we just go pitch them out onto the plant floor. Maybe there’s some quote, unquote mentorship that is happening during that process, but at some point, it just becomes, “Good luck, you guys will sink or swim.” And we completely ignore the technology that we have today. And for me, going and finding a way to blend those technologies is really interesting.

Chad Ghalamzan:

Switching topics a little bit because I think we could go and spend lots and lots of time about manufacturing and your passion for it, which is quite obvious, but there was something that you have for people who go and visit your LinkedIn profile. You have in your header three words that really drew my attention. I’d like to get more of an understanding from you why you put that there and what that means to you. “Sensible digital transformation” — it’s, like I said, right there on the top of your LinkedIn profile. First, what is sensible digital transformation, and why is it so important to be sensible?

Dave Griffith:

As I look at digital transformation in a whole, most of the time we talk about it is it’s the complete change of everything within organisations. It’s the digitization, it’s the data, it’s the making decisions, hopefully in real time based upon the data. It’s all of these things. And, Chad, as I go talk to end users about that, it becomes overwhelming. It’s like, “Okay, we’re going to do all of this, but the time it takes to do it might be 10 years, the cost might be three times what we make in revenue over the course of the year. How are we possibly going to go do this?”

And I think that folks, including myself who just talk about, “Hey, we’ve got to go through this digital transformation or this business industrial transformation,” I think it’s a miss. So, couple of years ago I started talking about sensible digital transformation. And I think sensible is important because if there’s not a reason to do the thing, and in most cases if there’s not a business reason, which both helps us on the manufacturing floor and adds to the bottom line of the organisation, it is a really difficult, if not impossible, conversation to have.

While it would be great for people to be technologists like us, and I assume folks listening to this show as technologists, we’re always looking for new interesting things, always looking for new interesting opportunities. At the end of the day, the business is a business. It has to make business sense. So for me, sensible digital transformation are processes that have the return on investments to pay for themselves, that are quick and that are going to have the business outputs. So when I talk about it, I’m like, “Yes, let’s go have the conversation on where we want to be in five to 10 years because that’s extremely important.”

If we don’t go map out what we want to be, I say to some clients, “When we grow up.” When we get to this next phase, if we don’t know what we want to be, we have no chance in going to be able to get there. So we go understand on where we want to get, then we go map our current state, where we are now to our future state, where we want to be in five to 10 years. And then from that point, we can go figure which building blocks, which fundamental things do we have to do.

And then my necessarily not-so-secret secret is we go pick something small and we execute it on extremely quickly and then we go show people that this can be successful, that it helps the business in the business sense and that it helps the operators and the folks on the plant floor in their sense as well. And if we get everyone on our team, it makes this process much easier than just someone on the executive level putting the fishtown and saying, “This is what we’re going to do. We’re not going to have any conversation about it.”

Speaker 3:

This episode of the Engineer Innovation podcast is brought to you by Siemens Digital Industry Software, bringing electronics, engineering, and manufacturing together to build a better digital future. If you enjoyed listening to this episode, please leave a five-star review to help get the word out about the show. For more unique insights on all kinds of cutting-edge topics, tune in to siemens.com/simcenter-podcast.

Chad Ghalamzan:

But do you think the companies that you talk to, or the people you talk to, are getting the math right in terms of the return on investment? Because digital transformation offers a lot of potential, and you’re right, it is an investment to go that way. Are you seeing, for example, a misrepresentation of the value? What do you think is the issue in terms of seeing the benefits for digital transformation?

Dave Griffith:

Sure. So I think many folks going and pitching and talking about digital transformation, we’re not necessarily talking about bottom-line returns, right? We’re talking about, “Hey, we’ve got to go through this digitization because we have to go through this digitization.” Very infrequently are we saying, “Hey, you guys currently, once a day or once a week, are going through these meetings to talk about what went right, what went wrong. We’re doing it on a piece of paper that is printed from an Excel spreadsheet that the data is three days old. We are saying, ‘Hey, this is a problem. Let’s go talk about how to fix the problem.’ But we all believe we go back to our offices, we set that piece of paper on a stack of paper, we probably don’t go think or look at it for two or three weeks, and then maybe two weeks or a month later, we actually solved the root cause of that problem.”

But that root cause of the problem cost us half a million dollars because we missed it for the days before and the weeks after we initially had that conversation. So I think very few people are talking about what the returns of digital transformation looks like because those numbers are difficult. And when you get to the point of going and talking about those business returns, you then generally need to bring the accountants and the CFO and the comptrollers in, and they may or may not know what the process looks like.

And there are a couple of handfuls of people that can go ahead and have those conversations of, “Hey, this is what we’re seeing, this is our calculations. Can you, as the accountants, as the CFOs, confirm this is the calculations that you are seeing?” And again, it’s a lot easier to say, “We’re going to go through a digital transformation and we’re going to add 10% to the bottom line, and that 10% in the bottom line is going to be five years from now.”

It is certainly possible that the business or the organisation appreciates 10% over the course of the next five years completely organically. And if we’re not going back and looking at tying them to specific projects, it is difficult. So what I have done is every time I go through a sensible digital transformation, when I look at projects, I do my very best to tie those back to return on investment to what we are seeing here in today.

And again, it’s difficult to project what’s going to happen in five years for anyone. I’ve got a friend who likes to talk about forecasts and the only thing he knows about forecasts are 50% of them are high and 50% of them are low. What we can do is we can say, “Hey, we’ve got this project. We’re going to do the project in the next three months. This is how we’re basing the project on. Let’s go follow this along.” And if we do find the 20%, if we do find the increase that we expect to find, then we’ve got the additional capital both in physical monetary terms of the business organisation to spend more on the next project we’re doing to reach the future state of the goal we have.

We’ve also got the capital of the people on the plant floor who have seen the work that we’ve done, makes their life easier, helps them with what they’re going to do. And hopefully, we’ve built internal champions on both sides so that when we get to phase two of what this looks like, we can go have slightly loftier goals, but we’ve got the people on every side who believe in the goal that we have for the company.

Chad Ghalamzan:

What’s a myth about digital transformation you’d like to dispel, then?

Dave Griffith:

The biggest myth that I’d like to talk about digital transformation is the fact that the way we talk about it, we snap our fingers and it’s done. Digital transformation is like artificial intelligence. I cannot purchase one. No one has a skew for one digital transformation that costs X number of dollars. Digital transformation is not a thing, it’s a process. It’s an organisational change. It is not easy, but it is possible. And if an organisation commits to it, it can truly change the entirety of the organisation.

Chad Ghalamzan:

So it’s not simply add to cart and checkout. But is it worth the transformation? Do you think the transformation people are unrealistic about the effort it takes or they’re unrealistic about the benefit it brings, or both?

Dave Griffith:

I think both, right? So I think that in almost all projects we underestimate the time it takes to go do so. We certainly underestimate the length and duration of a digital transformation because like most things, digital transformation means different things to every organisation. So if someone were to call me and say, “Hey, Dave, I’d like one digital transformation or I’d like one sensible digital transformation.” I’d be like, “That’s great. What does this word mean for you?” And, Chad, most of the time when I go ask that question, if I’m sitting in the boardroom, if there are 12 people, we get about 15 different answers.

So no one really knows what it is. So first step of defining what is success is to define what is a digital transformation, then to define what is success, and then to understand the process that we have to go through. So it’s very easy to say, it’s very easy to talk about in terms of marketing and other things. We certainly undervalue the time and commitment it takes for an organisation to go through. But if you go look that there are certainly a number of case studies that exist for large organisations who have either gone through a complete digital transformation or have done segments of these that find huge benefits as they go through these.

Chad Ghalamzan:

As someone coming from the manufacturing side, how do you see simulation playing a role in driving the digital transformation?

Dave Griffith:

I am very excited about simulation. Every time I talk to people about simulation, I think it’s amazing. We had the kind of point earlier of one, I built an aircraft facility and what that simulation looks like. And I think that there are some industries that do really well on simulation currently, like automotive does amazing on simulation, but automotive, they have lots of money for this. So they have lots of robots, they have lots of simulation and all of these things.

I am very excited as to what simulation can do for smaller facilities. When we get to the point of we’ve got small and medium-sized manufacturers leveraging simulation, I think it’s really exciting. One really good example, I worked with a mid-sized co-packer. They did all sorts of beer and seltzers. I worked with them last year, and they had this very good, but this reoccurring problem.

So a client would come and they’d be like, “Hey, I’d like to go run this slightly different recipe. I’d like to go run this new thing through the line. Can you confirm that it works on the line?” And these guys were so busy, it was really difficult to go take down the line long enough to go change temperatures and set points and go run it through the pillar and all of these other things. And at some point these guys are like, “I know I should do this, but we’re behind on production so I’m like 95% sure we can go run this.” So I’m just going to go say, “Yes, we’ll run it.” And if we’ve got catastrophic failure on the back end when we’re running, we’ll go figure it out.

So I’m over here and like, “Man, this would be a really good simulation opportunity.” We can go simulate the wine, we can go simulate everything that we’re going to go put inside of the drink, inside of the can, and we should have a very good idea as to if this will actually work or not. And by doing that, we don’t have to go take down the line, we don’t have to do the changeover and the setups and all of these things. And so, I am very excited as to what that looks like going forward, especially if we get into more high-mix environments.

So if we get into groups who are running lots of different products, and especially on the co-packer side, I see lots of groups running different products. This would give us the opportunity to go simulate, “Hey, will this work? What will happen if I do this?” I don’t have to go try to take down my line to go do a changeover for this. I can just go run it. And with some exceptionally high levels of certainty, we can go ahead and do that.

So I am exceptionally excited about simulation, especially for those small- to medium-sized organisations because that is where there’s huge benefit to the technology. We already have the big players in the market who I am very thankful of, continue to go promote and go find new use cases for it. So I think that when we get beyond the largest players to simulation and when we go have conversations around simulation.

I guess the first time I had a conversation around simulation and digital twin, I’m like, “Oh, that would be amazing, but that is hundreds of thousands or millions of dollars.” Then I go have conversations, and I’m like, “Oh, we can get digital twins of varying degrees, we can import CAD in other files, and it isn’t nearly as expensive as I had previously considered.” So I’m excited for continuing to have those conversations around simulation to go bring that technology to groups who would exceptionally benefit from it, but may think that it’s currently out of reach or they just technically are not capable enough internally to be able to use those tools.

Chad Ghalamzan:

Well, real data with the digital twin is obviously an important part of the digital transformation, and it sounds like your description from earlier is that some facilities haven’t kept up with the technology, and their lines are not necessarily running the right equipment to do that. But would you say it’s worth the effort, then, to leverage all the benefits of these different advances?

Dave Griffith:

In a word, yes. In a slightly longer than a word, I think that the vast majority of organisations will benefit from some, if not all, of these technological changes that we’re talking about. I think it is very much the conversation of how can we go through the process of going and finding the correct steps, again, the correct smaller projects to go bring this technology so that we can then leverage simulation and digital twin and all of these other things that we talk about.

Because going in and saying, “Hey, I would like to, again, one digital transformation, one digital twin, one simulation,” those are very large bytes, and it may go scare people away. So I think that the vast majority of these groups will get into more automation, will get more into the data capture and collection and hopefully understanding the realtime data.

And then once we’ve started walking down this path of industry 4.0 or digital transformation series of initiatives, then the next step is obviously going to become how can I go leverage digital twins? How can I go leverage simulations in my day-to-day? So a really good example, a project that I didn’t do, so I can go ahead and talk about on here, Chad, is I was working with a company that makes aluminium beer cans.

And so I was going and I was talking about a project, and I met with this group, and this was five or six years ago. So they knew similar to pharmaceutical and other groups today that they have a series of set points, and if they go hit those set points down their line, down their crosses, they’re going to get a perfect hand and they don’t have to worry about it. Right?

So for them, phase one is let’s go identify these set points, let’s go collect the data to make sure that we’re getting it. And if we’re not getting these, let’s go make changes as quickly as possible so we don’t lose a thousand cans a minute or however many large volume of cans that were going out.

But for them, the second opportunity was how can we go down the path of changing the set points in process? And I feel like we’re talking a lot about that today, and that is very much the digital twin with real time data. Then we’ve got a little bit of change in control. So if I know one set point is a little high, if I can go change the next set point down the line, I, in theory, could go save all of those in this case, cans, which is a huge value.

And so when you go talk with facilities that run very well or facilities that don’t run very well, the ability to understand that everything is going to be perfect or the ability to understand that, “Hey, we might have a problem, but we’re going to go change this in process to reduce that problem as much as possible.” That is the next level. That is the next huge breakthrough series of opportunities that we will certainly again see the largest of companies, well, if they’re not already doing it, hopefully they will start doing it.

And that will continue to trickle down as the important metrics become more and more important. And as we have the data and have the understanding of our processes to be able to say, “Hey, why are we having this 1% or 5% or 50% scrap rate?”

Chad Ghalamzan:

Yeah, change is important. Change is scary, and it can be expensive, but with simulation and digital twins, sounds like you can at least be a little bit more sensible about it. Something you’d agree with?

Dave Griffith:

Absolutely, yes. I think that change is certainly important. The only thing that we know in manufacturing are things will change. Everything about it is how we react to it. And these new tools, digital twin simulations will provide us with the best information as early as possible so that we can react and/or hopefully solve the problem before it gets very large.

Chad Ghalamzan:

Seems every guest we have on this podcast, artificial intelligence and machine learning comes up in one way or another. More specifically, do you think there’s an opportunity for AI and ML to help with the sensible digital transformation?

Dave Griffith:

Absolutely. I have lots of conversations about artificial intelligence and machine learning. I certainly think that they are some of the biggest buzzwords that we find in this industry. We are at the very nascent stages of actual machine learning and artificial intelligence. I think a lot of times we talk about artificial intelligence, and it’s just linear regressions.

And as I tell everyone in the super nerdy way I am about this, Chad, is that if we can go run the thing in an Excel formula and it can go give us the answer, it is not machine learning or artificial intelligence. Having said that, I think that there are a couple of very specific applications that we are seeing very good, exciting opportunities on. I think we’re seeing a lot of that, especially in machine vision and quality.

So too many of the facilities that I work with have very poor-quality setups, and lots of them are people hand-visually inspecting things and hand-visual inspection, while slightly better than nothing is only slightly better than nothing because we still have issues that come up. One of the interesting early opportunities that we will find with artificial intelligence is going and setting vision cameras up, and many times they don’t have to be very expensive vision cameras.

We’ll go set vision cameras up, we can go run a bunch of good and bad items through an untrained algorithm, and we can at some point very quickly go train that algorithm to understand what is good, what is bad, and then we’ve got a vision setup. And I think that that is really good for a couple of reasons. One, gives people the opportunity to go leverage these different technologies on the plant floor. It goes, helps these technologies move forward.

Two, vision setups like that, it is very much a combination of art and science in our industry. There is a bunch of science that says, “We can go ahead and do this,” but then there’s also a bunch of artistic, “Hey, the lighting has to be perfect, and do I have to go block out some of the other lighting or the skylights or other things, and why is this taking so long?”

And then I’ve got to go run hundreds or thousands or tens of thousands of images through a system so that it can understand yes or no. I think that there’s always a large lift, and so I think that we can go take some of those very monotonous human tasks and go run them through algorithms like AI and ML and go be able to successfully find solutions based off of that.

So I am excited about that. I think that, realistically, we’re a number of years away from large AI and ML offerings. I think we also go back and look at what sort of data are we getting from most of these facilities and from most of these facilities, if we do have data, it is almost certainly neither clean and/or as precise as we need to go, hand to a data scientist.

And then once we get to the point of that, I think that there will certainly be good opportunities. But again, there are so few facilities that can go hand the last six months or 10 years of data over to a data scientist that it doesn’t take six months to three years just to go through the process of cleaning all that up.

Chad Ghalamzan:

Yeah, data cleanup, data management is a big part of a system, like you mentioned. There’s also the potential to take that data and do more than just do the visual inspection, but then use it and bring it back to the design phase, bring it back to the manufacturing part and say, “Well, is this D flex occurring because of a specific point in the process?” And correct that.

So that continual life cycle, sharing of data for products helps with tools like artificial intelligence, to make sure that data is properly transmitted to the right groups. But you keep saying it’s a buzzword. Do you feel that there’s a lot of hype but not a lot of substance here? Or is it just the lack of clarity of all the ways it can be implemented and perhaps we haven’t discovered the full benefits of what AI will offer, not just manufacturing, but all the different aspects that can be applied to it?

Dave Griffith:

Absolutely. You’re going to get me in trouble here, Chad, but we will go for it anyway.

Chad Ghalamzan:

We like to get people in trouble.

Dave Griffith:

Perfect. In my opinion, and I should state this is my opinion, I think that we’ve got lots of buzzwords in manufacturing. I think industry 4.0, I think digital transformation, I think AI, I think ML, I think lots of these things are very much hype and buzzwords that marketers use and many times marketers use well, in order to go drive conversations around very topical things.

Again, being the practitioner that I am, I am more than happy to go have conversations around these topics, and I am as guilty as probably anyone to go drive conversations towards these topics because they’re the new fun, exciting things. But as the practitioner, I then go to the plant floor. And again, it’s 1954, right? So we are long ways away from going and fully implementing the vast majority of these things.

I think that we certainly need to go as technologists and practitioners go trial, how can we do this? So how can we go drive the most value that we possibly can, using the tools? But just like I can’t go buy one digital transformation, I can’t go buy one artificial intelligence. If someone knows where I can go buy one artificial intelligence, irregardless of the cost of the skew, I am more than happy to go purchase one to go trial it out.

Chad Ghalamzan:

Have you asked ChatGTP where you could go buy one to see what it says?

Dave Griffith:

I have not. And I’m really tempted to go launch it right now, Chad, but I won’t for the sake of this podcast and this show. But no, we can’t go buy one. So I think that as much as I am excited to go talk about this, I feel like it is the responsibility of us as practitioners to be able to go say, “Hey, this is the goal. This is the future state of where we’d like to be, but currently we are at X. If we want to go try artificial intelligence…” again, that vision application, the vision quality application, “Absolutely, let’s go ahead and trial that out. If we want to go trial connected workforce applications, absolutely let’s go trial that out.” If we want do other things that fall into industry 4.0 or digital transformation or any of these other very large buckets, that means so many things to so many people, we can absolutely go ahead and trial that out as soon as we understand what it means to the organisation.

And then we’ve got realistic, measurable goals and opportunities with these projects. Otherwise, these projects just go on for years and years, and at some point, the projects just continue for so long, it is almost impossible for everyone to be happy and for it to be successful at the end.

Chad Ghalamzan:

So it sounds like you more or less believe in the benefits these things will bring about, but the journey is a little bit more challenging than is portrayed from time to time. Is that a fair assessment?

Dave Griffith:

Absolutely, and I would agree with that. And I think that the journey is more difficult because, in reality, very few people actually go down the journey because these are journeys that are many, many years long, many projects long. And so the vast majority of people, again, that we go talk about all these things, we are not living that journey with the end user in the entirety.

So we’re doing little bits and pieces and those little bits and pieces are absolutely important, but it is all a journey. And so I think that we are very much still in the nascent stages of seeing what people started five, six, eight years ago back when industry 4.0 was the initial buzzword coming out of Germany. I think we are starting to see some of those values, and I think we see lots of really good application specific benefits from different companies who have put great offerings with that. But as we look at these things in a whole, it becomes that we need to go have a larger conversation as to what digital transformation one artificial intelligence actually means.

Chad Ghalamzan:

Dave, you co-host your own podcasts, so I hope this won’t put too much pressure on you. What’s the one question you wanted me to ask you today?

Dave Griffith:

From my perspective, Chad, one of the things that I’m most passionate about is how do people get there? So it is maybe we’ve got some technology, maybe we’re interested in some technology, but how do we get there? And it is probably one of the things that I’ve been most passionate about throughout the course of my career. And I did talk about going and taking those bite-sized pieces of any of these transformational journeys.

So I guess from my side, I’ve got a couple of points to that. Again, understand where you want to get as an organisation, go pull in the key shareholders and stakeholders. And I find that many people miss in the operations folks, the folks on the plant floor through the processes that I do, those guys are my heroes. If we’ve got good data, they know the actual pain points that correlate to that data.

If you don’t have data, as many groups don’t, the amazing part is, Chad, that those people still know what the pain points are and 95% of the time, their pain points correlates it to major downtime, major opportunities. So go understand where you want to be, and then, if you’ve got the opportunity internally, go have the conversations of, “Hey, how can we go build this internally?” Or, “Who do we go bring on to help us build this?”

I would say in the process that I have built kind of the workshops and other series that I do with organisations, take it a bit further. I’m happy to go ahead and outline it for anyone who might be interested. So what I do is I go look at a process map, right? I want to understand what the facility looks like. I think with operators as our heroes, the folks on the plant floor again, know what the issues become.

It’s a, I want to go ask them the questions of, “Hey, what are your pain points?” And I find that many of the pain points correlate to intersections between a number of different groups. And if we’ve got the intersection between two or three groups and it’s a pain point for all of them, it’s almost certainly a huge opportunity. And if we can go solve a problem for three or four different groups all at the same time, that is an amazing opportunity to go start down the path of a digital transformation of this transformational process and capture a bunch of champions and capture a tonne of goodwill from everyone you work with.

Chad Ghalamzan:

And what’s the one question you didn’t want me to ask you today? And I’ll leave it with you to answer it optionally.

Dave Griffith:

I’m always happy to go have conversations with people, so I’m generally very open when it comes to this. I guess the one question I didn’t want you to ask today, Chad, are for very specific recommendations on very technical problems. Those are always the most difficult to go ahead and answer because there is almost never a one-size-fits-all solution to a very particular thing.

Chad Ghalamzan:

Well, that’s some wise words to leave today’s episode on. Dave, thank you for your time. It’s been a fascinating conversation. I hope you’ve enjoyed it as much as me.

Dave Griffith:

Absolutely. Thank you so much, Chad. It has been a fantastic time. Very happy to have this conversation. Thank you for having me on.

Chad Ghalamzan:

And thank you for listening to the Engineer Innovation podcast, powered by Simcenter. If you’ve enjoyed this episode, please leave a five-star review and be sure to subscribe so you never miss an episode. My name’s Chad Ghalamzan and I’ll speak to you next time.

Speaker 3:

This episode of the Engineer Innovation podcast is powered by Simcenter. Turn product complexity into a competitive advantage with Simcenter solutions that empower your engineering teams to push the boundaries, solve the toughest problems, and bring innovations to market faster.